��ɫ��Ƶ

Links

Key Discussion Topics & Highlights

❤️‍🩹Definition of Health Data

• Health data includes patient records, doctor/nurse notes, NHS numbers, date of birth, data from tissue samples, and wearable fitness trackers.
• Not all health data is stored in one place; it is fragmented across different sites and systems.

⚖️Access and Ethics

• Researchers must apply for access to health data with clear ethical justifications.
• Multiple layers of approval and consent processes exist to protect privacy and ensure responsible data use.

🚇Concept of ‘Data Journeys’

• Rather than linear “data flows,” Itzelle prefers the term “data journeys” to reflect complexities, delays, and ethical checkpoints involved in data movement.

🙈Public Attitudes toward Data Use

• People generally support health data use within the NHS and for public good (e.g. research, policy-making).
• Trust diminishes when external commercial entities are involved (e.g. Google DeepMind).
• Lack of transparency leads people to fill in the blanks with imagined negative scenarios.

🤔Concerns about Profit and Misuse

• Concerns arise when profit is a motive—especially around selling data or influence over care decisions.
• Vulnerable individuals (e.g. migrants, LGBTQ+ individuals) may fear government surveillance or discrimination.

⌚Wearables and Wellbeing Apps

• People use digital health trackers for self-monitoring, but reliance decreases over time.
• Attitudes vary: some feel lost without their data, while others are indifferent.
• Over-reliance on wearable data for policy-making may exclude those without access to such technologies.

👓Need for Transparency

• Being open about data sharing practices and involving the public in decision-making builds trust.
• Transparency is especially crucial when external partnerships or new technologies are introduced.

🔬Current Research Projects

• Patterns in Practice: Studying how AI practitioners' values influence their work in education, pharma, and the arts.
• AI in the Global South: Exploring public expectations of AI for solving societal issues, with a focus on Latin America.

Episode 2 Transcript

Hannah Clemmens (HC): Hello and welcome to the Digital Health Hubcast, where technology and healthcare collide. I'm your host, Hannah Clemmens from the South Yorkshire Digital Health Hub, where we're on a mission to tackle health inequalities by transforming the future of digital health. Digital innovation is at the heart of the future of healthcare, and on this podcast, I sit down with leading experts and innovators from across the digital health landscape to discuss the biggest opportunities, toughest challenges, and how we can innovate in this space to create a healthier and more connected future.

In the spotlight today is Itzelle Medina-Perea, a lecturer in Information Systems at the University of ��ɫ��Ƶ. Originally from Mexico, Itzelle has been living in the UK for almost 10 years, where she's been exploring the factors that shape data flows, data practices, and the uses of data in new and emerging technologies. Her PhD research contributes to understanding how different factors work together to shape the movement of health data within the UK healthcare sector.

Itzelle has been involved in two large-scale projects – Patterns in Practice, which explores how practitioners’ beliefs, values, feelings, and emotions interact to shape how they engage with data, and Living with Data, which explored how people, the public, and patients feel and think about data uses within different domains of everyday life.

Itzelle, welcome! It's a real pleasure to have you here today. How are you doing?

Itzelle Medina-Perea (IMP): Thank you, I'm really good. Yeah, thank you for having me.

HC: Lovely. Yeah, I'm really excited to talk a bit more about your research on how data flows from the NHS or other health data sources into academia and how people feel about that. So, I'd like to start by really talking about what do we mean when we say health data?

IMP: Okay, yeah, so this is an interesting question that has a short answer but also a long answer, so I think I'm going to go for the long answer this time. So, health data is basically data about any person that is connected to their health, and it's basically about their health conditions, their quality of life—so it's about these things. We can find it in different formats or in different places.

So, we can have health data in patient health records, in studies about the health of a specific group of people, in data from blood or tissue samples, and data also from fitness devices—so these trackers that people sometimes use. We have health data in these different ways and in my research, in my PhD, I focus mainly on patient data. So, patient data basically can include data about a medical condition but also notes that a doctor might take in an appointment about you. But not only a doctor—it could be a nurse or any other health professional. It also includes your personal details like your NHS number, your date of birth—so basically, all of that can be contained in patient data records.

So yeah, that’s kind of the long answer of what do we mean by health data.

HC: Yeah, yeah, and I'm assuming this is all stored somewhere very securely? Can you tell me a bit more about that?

IMP: So that is also not a straightforward answer. I think that if we try to imagine where—like, if you asked me before this research what I thought about this—I would say, “Oh, probably it's just stored all in one place and all the data about me is in a single record online somewhere.” Actually, it's more complex than that. So, in the NHS, the data can be stored in different places and it's very messy and very difficult to track.

Some data stays only with your direct provider, whereas other types of data can go to other places. So, we can say that NHS data is stored in different places—or health data, in the case of the UK—is stored in different places, and it is safe and secure.

HC: Who might have access to that?

IMP: So that also depends. You can request access to your data as a patient, and you can also for some specific parts of your health record. If someone else wants to have access, you need to give that permission. That doesn't mean that you need to give permission every time another person uses the data, but maybe you might remember when you register with a GP, sometimes they give you a form and then you have to give consent if you want to give access to other people for other purposes. So, people that can access this data include obviously the people that care for you, like the care providers—so people in the health sector—but also researchers might access that data to explore health conditions about a group of people. You can give access also not to your main provider but to other services that are other health services.

HC: So what does that look like then? If I'm a researcher and I want to study health data, how do I access that?

IMP: So, if you are a researcher and if you are based in a university, for example, there are different ways of accessing data. So, one of those ways is that you—well, first—you have to have a question in mind. You need to have a purpose for why you want to access the data. You need to approach the specific site that holds the data. For example, I'm going to give a concrete example because I think otherwise it's very ambiguous. So, for example, one of the cases that I explored during my research, a group of researchers in Manchester wanted to access data about antibiotic prescriptions. So they had to approach the specific sites or the specific clinics where they wanted to get access to data from. They had to explain what the reason was, and then they had to do a lot of paperwork. So, it's not that easy for people to access data—because it should be protected.

But yeah, you need to approach the site and then go through a process where they need to make sure that the data is going to be used in an ethical way and that it is going to be managed appropriately. You can also access data through other—like, in different ways. There are some places where there are repositories that hold data at a national level, so then that's again a separate process. If you want to do a piece of research, you might need to get access to different pots of data, so it's like going to different places and obtaining different permissions to access that data.

HC: Is that because it's all in different places?

IMP: It's in different places, yeah, so that's why you need to go to different people to meet different requirements.

HC: It sounds quite complicated.

IMP: Yeah, it is. It can be.

HC: So in your research, we're obviously talking about data and how it flows from health sources into research. So I'm wondering if you can tell me a bit more about data flows—you use that term "data flows" in your research. Can you tell me a bit more about what you mean by that?

IMP: Yeah, so data flows refer to the way in which data moves from one place to another. So that's the definition of flow. In my research, I adopted another term, drawing from research that other people did in the past. So I'm using the concept—rather than talking about simply data flows, I'm talking about data journeys, which is also a way to talk about the movement of data.

However, it has some distinctions, and that’s why I prefer to talk about data journeys. This concept of data journeys was proposed by a group of researchers, and the lead of that research group is Jo Bates, who is also based here at the University of ��ɫ��Ƶ.

The term "data journey" also talks about how data moves from one place to another, but this analogy of a journey makes you think and realise that it's not just data moving from one place to another in a very easy way. When you think about a journey, you think that there are stops, there are some moments where there is a pause, or there are moments in which you can go faster or slower. So that’s why the term “data journey” comes into this research. It’s very messy, so it's not just so easy.

HC: Yeah, yeah, that makes sense. And those kinds of obstacles—are they to make sure that the use of the data is correct, it's ethically sound, or are there other reasons?

IMP: Yes, these obstacles can also be called frictions—so yeah, those things that prevent data from moving so easily. Sometimes they have been referred to as a negative thing in that they prevent data from moving, but actually, if we consider that this is very sensitive data and that it should be protected and managed carefully, those obstacles become a positive thing because it's a way of protecting how this data is used and who gets access to it.

HC: Yeah, yeah, of course. That leads me really nicely on to my next question, which is to speak a bit more on who else this data is valuable to. So it's clear how it's important for researchers to do research on health and hopefully cure diseases, but also—yeah—who else might benefit from this data?

IMP: Yes, as I was mentioning, obviously the first beneficiary is the patient—or if we talk in plural, well, the patients—all of us that receive a health service. And then, yeah, we have researchers that can use this data to investigate a specific issue. But also, this data is valuable for healthcare planning because it's valuable for creating policy. It’s very helpful to make informed decisions. It can also be helpful to develop health interventions or to develop different ways of assisting people in things that are connected to their day-to-day health.

HC: So it's that kind of bigger piece of looking at lots and lots of people's health data that can inform policy, like you say—is that what you mean?

IMP: Yes.

HC: That makes sense. Let's talk a bit more about your research on how people feel about the movement of their health data. You worked on a project called Living with Data—is that right?

IMP: Yes, yeah.

HC: Why don't you tell me more about that?

IMP: Yes, Living with Data was a project in which we wanted to understand how people feel and think about the uses of personal data in different sectors. This research was also conducted by a group of researchers based here at the University of ��ɫ��Ƶ. This was a project led by Helen Kennedy, who's based in Sociological Studies, and then the group came from other places in the University.

We wanted to understand how people feel about the uses of data—not in general, but in a very specific context. So, we looked at the uses of personal data in the public sector. We looked at health data in the NHS, the uses of personal data in the BBC, and the Department of Work and Pensions (DWP).

As part of this research, we conducted surveys—two waves of a survey—to understand how people feel about this. But we also included focus groups and interviews to have more detailed conversations with people about how they feel and think about the uses of data.

HC: What did you find?

IMP: We found many interesting things, and I think I'm going to focus on some of the key areas. Obviously, because today we are talking about health data, I'm not going to go into detail about the other cases, but focusing on the health data.

Because we wanted to understand how people feel, we also wanted to give them concrete examples of real uses of personal data—or of their health data. So in the surveys, and also in the interviews and focus groups, we shared with people two examples.

One example was about the use of data to inform antibiotics prescription. For this, we drew on my PhD research, which I just talked to you about. And then, because our research was conducted during COVID, we also wanted to bring a case that was meaningful in that context, so we also gave people one example about the COVID data store—the NHS COVID data store.

We asked people—giving them a brief explanation of these uses of data—and then in the survey we just asked the questions, but in the interviews and focus groups we asked more detailed questions. We presented visualisations to people to show them how data moves from one place to another.

We found that people tend to trust the uses of health data when they happen within the NHS. They were very open to this and also very comfortable when data is used for the good of everyone. That is one key finding from this research—that the context matters a lot. It's not only that people support the use of patient data within the NHS, but they support the use of data when it is for the benefit of everyone. So that’s one of the key findings from this research.

People are more concerned when they see the involvement of external stakeholders. Even when data is being used within the NHS, if there is an external provider involved—for example, providing infrastructure—people tend to feel concerned because of the potential consequences that this might have. So that’s one of the key findings we found in the research.

HC: Yeah, yeah, that makes sense. And those kinds of obstacles—are they to make sure that the use of the data is correct, it's ethically sound, or are there other reasons?

IMP: Yes, these obstacles can also be called frictions—so yeah, those things that prevent data from moving so easily. Sometimes they have been referred to as a negative thing in that they prevent data from moving, but actually, if we consider that this is very sensitive data and that it should be protected and managed carefully, those obstacles become a positive thing because it's a way of protecting how this data is used and who gets access to it.

HC: Yeah, yeah, of course. That leads me really nicely on to my next question, which is to speak a bit more on who else this data is valuable to. So it's clear how it's important for researchers to do research on health and hopefully cure diseases, but also—yeah—who else might benefit from this data?

IMP: Yes, as I was mentioning, obviously the first beneficiary is the patient—or if we talk in plural, well, the patients—all of us that receive a health service. And then, yeah, we have researchers that can use this data to investigate a specific issue. But also, this data is valuable for healthcare planning because it's valuable for creating policy. It’s very helpful to make informed decisions. It can also be helpful to develop health interventions or to develop different ways of assisting people in things that are connected to their day-to-day health.

HC: So it's that kind of bigger piece of looking at lots and lots of people's health data that can inform policy, like you say—is that what you mean?

IMP: Yes.

HC: That makes sense. Let's talk a bit more about your research on how people feel about the movement of their health data. You worked on a project called Living with Data—is that right?

IMP: Yes, yeah.

HC: Why don't you tell me more about that?

IMP: Yes, Living with Data was a project in which we wanted to understand how people feel and think about the uses of personal data in different sectors. This research was also conducted by a group of researchers based here at the University of ��ɫ��Ƶ. This was a project led by Helen Kennedy, who's based in Sociological Studies, and then the group came from other places in the University.

We wanted to understand how people feel about the uses of data—not in general, but in a very specific context. So, we looked at the uses of personal data in the public sector. We looked at health data in the NHS, the uses of personal data in the BBC, and the Department of Work and Pensions (DWP).

As part of this research, we conducted surveys—two waves of a survey—to understand how people feel about this. But we also included focus groups and interviews to have more detailed conversations with people about how they feel and think about the uses of data.

HC: What did you find?

IMP: We found many interesting things, and I think I'm going to focus on some of the key areas. Obviously, because today we are talking about health data, I'm not going to go into detail about the other cases, but focusing on the health data.

Because we wanted to understand how people feel, we also wanted to give them concrete examples of real uses of personal data—or of their health data. So in the surveys, and also in the interviews and focus groups, we shared with people two examples.

One example was about the use of data to inform antibiotics prescription. For this, we drew on my PhD research, which I just talked to you about. And then, because our research was conducted during COVID, we also wanted to bring a case that was meaningful in that context, so we also gave people one example about the COVID data store—the NHS COVID data store.

We asked people—giving them a brief explanation of these uses of data—and then in the survey we just asked the questions, but in the interviews and focus groups we asked more detailed questions. We presented visualisations to people to show them how data moves from one place to another.

We found that people tend to trust the uses of health data when they happen within the NHS. They were very open to this and also very comfortable when data is used for the good of everyone. That is one key finding from this research—that the context matters a lot. It's not only that people support the use of patient data within the NHS, but they support the use of data when it is for the benefit of everyone. So that’s one of the key findings from this research.

People are more concerned when they see the involvement of external stakeholders. Even when data is being used within the NHS, if there is an external provider involved—for example, providing infrastructure—people tend to feel concerned because of the potential consequences that this might have. So that’s one of the key things, well like kind of the main thing, that we found in the research.

HC: Who do you mean by an external partner?

IMP: So, um, here I'm talking about an external private company. For example, if there is a company providing the infrastructure—like let's say Google DeepMind or Amazon Web Services—people don't feel comfortable about that. And I mean, that's understandable, also because we have had in the past some cases where this type of external stakeholder has not been very straightforward with the ways in which they use data or in the ways in which they access the patient data. So that's understandable.

HC: What do you mean by that—so they could see it, they could share it—what do you mean?

IMP: So, um, one of the concerning cases was, I think, in 2019. One of these cases was when one hospital in London shared data with Google DeepMind, but they kept this very secret until it came to light thanks to an investigation. So it was not clear where this data was stored, who was accessing that data, or for what purposes. Obviously, if you are not very clear and transparent with these things, then a lot of questions can be asked. So yeah, that's what I mean.

HC: So I suppose it's important then for providers to be really clear about who they're sharing their data with or if they are using external partners—is that right?

IMP: Yes, yeah. And also, if they are using an external partner without telling people—so yeah, it's not only about being open about what they are doing, but it's also important to ask people how they feel about this before doing it. So yeah, that's one of the key findings.

Also, one very interesting thing is connected to these concerns that people have. In the interviews and focus groups, we found that when this information is missing—when you don’t know exactly how a provider, or how the NHS, or any other stakeholder is using data—people tend to fill those gaps of information with their own imagined scenarios. I'm not sure if I should use the word stakeholder, but you know, like the people handling the data.

If people don't have a clear explanation of how this is done, they can think about scenarios that are not really good. I mean, for example, in the context of the NHS data store—the COVID data store—the NHS kind of revealed the information little by little. On one side, that was understandable because everything was very complicated at that time. But when we had these interviews and focus groups and we didn’t know a lot about what was happening with the data store, some people were imagining things like, ‘Oh, maybe they are using this data to force people to get the vaccine,’ or, ‘Oh, maybe they are using this data to target a specific group.’ So people tend to use their imagination to fill those gaps. Obviously, it’s always better to be very straightforward with how things are handled.

HC: So just based on your own assumptions or beliefs, you kind of fill that gap with something that you’ve pulled out of nowhere—is that what you mean? But this isn't conscious, you mean?

IMP: Yeah, yeah, yes. So that was also very prominent in the focus groups and interviews we had.

HC: Wow, so transparency then can help with that?

IMP: Yes. And also, another thing I just want to mention—connected to these concerns—we also found that people feel differently about data uses depending on their own lived experiences. So people who are in vulnerable groups might feel more concerned than others.

We are not saying that it's because they are vulnerable that they think differently, but because their lived experiences inform their thinking, and that's why they feel differently about data uses.

HC: Yeah, that makes total sense. So would you say that people in those groups are more wary of how their data is being used?

IMP: Well, because they have different experiences in life, that’s why their interactions with the services are different. But what I want to say is that it is justified—so it’s not coming out of nowhere, it’s not imaginary.

HC: Yes, that makes total sense. So you’re saying then that generally, if it comes from a healthcare provider like a doctor, or if it’s for the greater good—if you like, so it’s for research or to make policy or to make wider healthcare changes—generally people are supportive of that. Yes? But when you start to involve external companies to provide services, or when there’s less transparency about what that actually means and how Google, for example, are involved, that’s when people start to feel uneasy?

IMP: Yeah, and also when there is the potential for profit, that also causes some concerns. And it's not that people are always opposed to profit, but they think that when that comes into play, then it's more difficult to make a decision—should we support this or not? If this company wants to make a profit, then that means that will be the priority. And then, how do we ensure that the data is protected?

HC: Yeah, yeah. When you say profit, do you mean making a profit from providing a service or making a profit by selling data?

IMP: It could be both things, but I think that probably the most concerning thing for people is selling data or using it in ways that—yeah, selling data for purposes that they don't even understand.

HC: So what would be the consequences of that? You know, I'm thinking of when I just see some random emails in my junk box and I think, "Where have you got my email address from?" Like, has this been sold somewhere? Is that what you mean, or is it other things?

IMP: So that can be one thing. But also, one of the concerns that people have is, like, what if this data is then shared with an insurance company or something like that, and then that affects the way in which I am perceived by these types of providers? Or what if they are using this data to make decisions about the population in a bad way?

HC: In your report—which we can link to in the notes, actually—something that really struck me: you have a few testimonials from people that you spoke to, or quotes from people that you spoke to, and one that's really stuck with me is a man who is a migrant to the UK. So, where he grew up back home, it's illegal to be gay, and as a gay man, he was really concerned that his data would be leaked to the government and then he'd really be in trouble.

IMP: Yeah, yeah. What you just mentioned is a really good example of this point that I made earlier about how people who are in a vulnerable position inform their thinking about the uses of data based on their own experiences. So this person experienced something in their country, so now they know what could be the potential consequences of that.

And here in the UK, like around 2015/2016, some health data was shared with the Home Office to identify people that were in the country without documents. So that's also—like, these concerns come from a reason, I guess.

HC: Yeah, of course, that makes total sense. You mentioned right at the start when we were talking about health data and you were sort of citing all the different sources that health data could come from, and you mentioned wearable data and smartphone data. So I'm wondering if we can talk a little bit about that. Digital health is here, isn't it? We all have apps and things to measure, I don't know—weight or drinking or something like that. So I'm wondering if you've learned anything about how people feel about that data. Because you're right, you could class it as health data, but when I think about health data, I just think about my health records. So yeah, I'm wondering if you can speak on that?

IMP: Yeah, so there has been research also exploring how people feel about this type of data and people that use trackers. Here, I’m talking about the research of a colleague also based at the Information School, Lee Pretlove, who explores how people feel about these data trackers or wellbeing trackers.

One of the things is that people use this data when they are at the beginning of making changes for their health. For example, they start running, so they get these trackers and they use this data and they rely on this data a lot at the beginning when they want to know how they are doing—if they are going too fast, if they are going too slow, how their heart is reacting to this.

But then, something interesting is that as time passes, they rely less on that data and they rely more on their body—on how they feel—and they understand that sometimes this data does not give the full picture. So that's one thing about this type of data.

Some people—and this also depends on their personalities—some people enjoy the act of collecting this data and getting to know more about themselves. When they are asked—and this is again drawing from that research from this colleague—how they would feel if they suddenly lost this data, some of them feel like, “Oh well, I would feel lost because this is like years of tracking how I felt over a long period of time.” But other people feel like, “Well, if I lose that data, I guess I’ll just carry on doing what I do.”

So yeah, and there have been suggestions that this type of data—like data collected from these monitoring devices—could be used to inform changes in the healthcare sector or inform policy. But one of the critical things to think about in relation to this suggestion is that some people use these devices, but not everyone. And it’s not always because of a personal decision—it’s also because not everyone has access to this type of technology. So, relying on that type of data, on one hand, could be helpful to inform, but also, on the other hand—who are we leaving outside if we decide to rely on that data that not everyone has access to? So it could drive an even greater divide.

HC: So tell me a bit more about what you're looking at now in your research. What's coming up? What's in the future?

IMP: Um, so right now I'm involved in two projects that are also connected to how people feel and think about the uses of data and technology, but in other contexts. I'm involved in a project called Patterns in Practice, in which we have explored how practitioners working with artificial intelligence—how their values, beliefs, and feelings—how all these things influence how they use artificial intelligence in their work. We are looking at three contexts: higher education, the pharmaceutical sector, and the arts.

Also, I'm working on another project in which we are exploring what expectations people have in relation to the use of artificial intelligence for addressing societal issues, but in the Global South. So, I'm looking at what is happening in Latin America. I'm currently working on these two projects for now.

HC: Amazing, amazing. I'm sure that's enough to keep you busy—that sounds fantastic. How can people find out more about you, Itzelle?

IMP: Uh, well, I have my University web page, so you can find a brief summary of what I do and my main interests and some of the outputs of my research that we just discussed. I'm on LinkedIn and—I mean, I have a Twitter account, but I no longer use that one. I'm planning to move to BlueSky.

HC: Yeah, making the jump.

IMP: Yeah, so I'm in the process. But for the moment, my University website and my LinkedIn account, yeah.

HC: Perfect. It's been such a pleasure to speak to you about how health data flows and what people feel. It's been a real treat, so thanks so much.

IMP: Thank you very much.

HC: I really enjoyed that talk with Itzelle. Here's my take on our conversation. Health data is held in various different places and isn't necessarily straightforward to access—and rightly so. It was reassuring to hear and learn about all the different barriers to access that are in place to ensure that data is managed correctly and only accessible to people doing ethically approved research.

It sounds like citizens and the public are generally happy to share their data so long as it's in the public good, but begin to grow concerned once external parties are involved. So for anyone handling people's data, it seems like transparency is really important to retain public trust.

You can find links to Itzelle's University profile and some of the research that's mentioned in the notes. I really encourage you to check out the Living with Data report—there are some really nice stories in there that I certainly enjoyed reading.

Thanks so much for listening, and see you next time.